Post on 20-Jan-2023
WorkingPaperWorkingPaper
ISTITUTO DI RICERCASULL’IMPRESA E LO SVILUPPO
ISSN (print): 1591-0709ISSN (on line): 2036-8216
Co
nsi
gli
o N
azio
na
le d
ell
e R
ice
rch
e
cover new impa ceris 2010 26-01-2010 7:36 Pagina 1
Marini G., Working Paper Cnr-Ceris, N° 03/2014
Copyright © 2014 by Cnr-Ceris All rights reserved. Parts of this paper may be reproduced with the permission of the author(s) and quoting the source.
Tutti i diritti riservati. Parti di quest’articolo possono essere riprodotte previa autorizzazione citando la fonte.
WORKING PAPER CNR - CERIS
RIVISTA SOGGETTA A REFERAGGIO INTERNO ED ESTERNO
ANNO 16, N° 3 – 2014 Autorizzazione del Tribunale di Torino
N. 2681 del 28 marzo 1977
ISSN (print): 1591-0709
ISSN (on line): 2036-8216
DIRETTORE RESPONSABILE
Secondo Rolfo
DIREZIONE E REDAZIONE
Cnr-Ceris
Via Real Collegio, 30
10024 Moncalieri (Torino), Italy
Tel. +39 011 6824.911
Fax +39 011 6824.966
segreteria@ceris.cnr.it
www.ceris.cnr.it
COMITATO SCIENTIFICO
Secondo Rolfo
Giuseppe Calabrese
Elena Ragazzi
Maurizio Rocchi
Giampaolo Vitali
Roberto Zoboli
SEDE DI ROMA
Via dei Taurini, 19
00185 Roma, Italy
Tel. +39 06 49937810
Fax +39 06 49937884
SEDE DI MILANO
Via Bassini, 15
20121 Milano, Italy
tel. +39 02 23699501
Fax +39 02 23699530
SEGRETERIA DI REDAZIONE
Enrico Viarisio
e.viarisio@ceris.cnr.it
DISTRIBUZIONE
On line:
www.ceris.cnr.it/index.php?option=com_content&task=section&id=4&Itemid=64
FOTOCOMPOSIZIONE E IMPAGINAZIONE
In proprio
Finito di stampare nel mese di Marzo 2014
Marini G., Working Paper Cnr-Ceris, N° 03/2014
Italy’s New Requirements for Academic
Careers: The New Habilitation
and its Worthiness
Giulio Marini
Cipes, Centro de Investigação de Políticas do Ensino Superior, Universidade do Porto
Rua Primeiro de Dezembro, 399 Matosinhos, 4450-227, Portugal
CNR-CERIS National Research Council of Italy, Institute for Economic Research on Firm and Growth
Via dei Taurini 19, Roma, Italy
Mail: gmarini@cipes.up.pt
Tel: +351 229 398 790
ABSTRACT: The new habilitation, established in Italy in 2010 and commenced in 2012, was
designed (outcomes released commencing December 2013). Its aim is to filter who will be eligible to
apply for competitions for the two permanent level professor positions in the universities. The results
of the first set of data are 20 scientific sectors representing more than 10% of all sectors analyzed to
understand if the outcomes reflected in a worthy way the indicators of productivity and quality of
scientific production of candidates. Some legal and statistical framework are fostered before the data
analysis in order to have a better understanding of the reform and the context where it operates. The
hypothesis of the worthiness is here addressed on the assumption that the current position held by a
candidate should not play any role in the attainment of the habilitation. Splitting candidates into two
roles and having controlled for age as a variable, the data was used to reveal that the indicators of
quality of scientific production (H index for hard sciences and articles in top ranked journals for
social sciences and humanities) are more frequently the best predictors. Though some limits of the
present analysis are faced and illustrated, some critical points of this new institution are discussed.
Keywords: habilitation, academic career, scientific productivity, rigged competitions, epistemic
communities
JEL Codes: J44
Marini G., Working Paper Cnr-Ceris, N° 03/2014
4
CONTENTS
1. Introduction: A system in the Pursuit of New Balances .......................................................... 5
2. Methodology. The Construction of a Unique Dataset: Methodological
Explanations ............................................................................................................................. 8
3. The Employees Pyramid of Italian Universities .................................................................... 10
4. Data Analyses: Were Committees a Source of Meritocracy? ................................................ 13
5. Conclusions and Discussions About Habilitation .................................................................. 22
References ................................................................................................................................... 26
Appendix ..................................................................................................................................... 28
Marini G., Working Paper Cnr-Ceris, N° 03/2014
5
1. INTRODUCTION: A SYSTEM IN
THE PURSUIT OF NEW BALANCES
ecruitment in Italian Universities.
The Italian public administration
started to face the problems
associated with public expenditures beginning
in the early 1990s. One of the key points at
those times was to take control of the
retirement system of employees and the
recruitment efforts of new personnel in public
administration positions. As such,
universities, however, did not make an effort
with respect to this initiative (Cavalli &
Moscati, 2010), even though the
implementation of ad hoc policies were not
designed permanently. Since that time,
though, a general assumption of frozen
organization charts was established to
regulate the systems. Rules and procedures of
competition (concorsi) have often changed;
sometimes even with decrees that were
designed to change the mechanisms (Degli,
Esposti, & Geraci, 2010). Nonetheless,
besides financial sustainability, the political
end has often been focused on efficiency and
meritocracy. Examining previous decades of
an open door policy of recruitment with a
subsequent long period of a lack of
competition, resulted in a juridical comparison
similar to the French physics context
(Pezzoni, Sterzi, & Lissoni, 2012) and ended
with a demographic misbalance along with a
reduction of productivity due to a lack of any
relevant filters in the recruitment process.
Now, as a result of this previous process, to
enter the university’s professoriate system
with a permanent position has become a step-
by-step process that is more difficult and, at
the same time, even more uncertain. If in
previous decades it was a matter of certainty
that to enter with a tenured position was only
a matter of time, now there is one more filter
(the habilitation) that must be overcome. In
any case, the previous tenure mechanism was
more or less always the same following a
traditional pattern: the maestro used his or her
personal influence to give the post to his /her
alumno, or pupil (Fassari, 2009; Pezzoni et
al., 2012; Vaira, 2011). In a more formal
way, it could be argued that the personal
influence described by Clark (1986) has
always played a crucial role along with the
current juridical mechanisms of concorsi.
These complex rules established by the
Napoleonic pattern were in many cases
plagued with allegations of rigged
competitions. Moreover, disciplines, rather
than institutions, continued to play a strong
role (Becher & Trowler, 2001; Lissoni,
Mairesse, Montobbio, & Pezzoni, 2011).
Despite the rumors and allegations about this
pattern of donship and nepotism, a different
sort of analyses had already been pursued
from different perspectives focusing just on
recruitment procedures and its possible
aberrant practices (Rossi, 2012b; Nelken,
2009; Paris, 2005).
Under this scenario, the 240/2010 Law, also
known as Gelmini Law, was established and
individuals tried to change this existing
paradigm. Gelmini Law contained harsher
new conditions that made initially entering
universities with a permanent position a
tougher goal. Under this new law, the levels
of position were reduced from three levels,
full professors, associate professors, and
assistant professors and replaced with only
two levels, full and associate levels.
Although the assistant professor positions
may still be given, they are now only able to
be awarded as fixed-term positions whose
renewals (typically 3 plus 2 or 3 plus 3 years,
respectively) should be tougher to maintain
R
Marini G., Working Paper Cnr-Ceris, N° 03/2014
6
and is no longer considered simply a formality
(previously passing from one level to another
was only a question of time or seniority, and
the amount of needed time could have been
assumed as a proxy of personal success and
influence).
Now, these positions as assistant professors
are expected to be even externally funded
and/or based on national competitions.
Unfortunately, until now the total amount of
competition funds like FIRBs, and recently
replaced SIRs, (both national research
competitions for junior professors who are
less than 40 years old) was not considered as a
serous incentive when compared just to the
mere necessities of renovations of the human
resources.
The Habilitation in Italy. The habilitation is
now being established as a new institution in
the Italian national context. Yet, habilitation is
not a new mechanism in academic
recruitment. It has been in practice for a
number of years in Germany, where
Habilitation is closer to what is now the new
fixed-term assistant professorship.
In that country, this is essentially a second
PhD and is usually over a different topic from
the original PhD’s dissertation (Enders, 2001).
However, that version is very different from
the Italian use of the term.
More recently it has also been implemented
in France (habilitation; Musselin, 2004) and is
now being introduced in Spain (abilitación –
that version had thresholds in order to actually
pre-indicate who would probably have
attained the positions –, more recently
replaced by acreditación, that has no
thresholds and looks to be more similar to the
Italian case).
Despite these similar terms, these systems
are enacted very differently within each
country and there is a lack of space for
comparison within the context of this article.
In Italy, the goal of this system is to provide
the possibility for any candidate to become
eligible to participate in future competitions
(concorsi) for full professorship (first level)
and associate professorships (second level)
without even the attainment of idoneità
(eligibility, or fit-for-the-job), a very common
way to give positions to losing candidates of
previous competitions with good scores.
Being able to successfully compete in this
kind of competition is a new hurdle to be
overcome and whose successful output does
not necessarily guarantee any sort of post. In a
more formal way, habilitation could be
described as a “tougher pool of candidate and
selective examination” because there is one
more step when compared to the
competitions, thus making this recruitment a
longer (in time) tournament (Musselin, 2004).
At the same time, the process also looks to
introduce some of the characteristics of the
opt or out mechanism, especially at the
associate level (Musselin, 2004), since people
who have not a permanent position are
compelled to get this habilitation in order to
survive1.
The evolution of the system needs more
time to be completed and eventually more
investigations to be conducted.
1 A probable example of this for the next years to
come could be represented by fixed-term assistant
professors who will not have all the possibility to
enter as permanent. Since their positions will expire
and they will be by that time around 40 years old or
more, a relevant problem of employment will be to be
overcome. Even for this, it can be hypothesized that
actually the system could favor “semi-insiders”
(people who are employees, even though not
permanently) toward outsiders (people who are
employees somewhere else or have less priority in the
informal queue of recruitment, such as for post-docs).
Marini G., Working Paper Cnr-Ceris, N° 03/2014
7
Despite the manifest intention to introduce
more meritocracy2 to the system, in a widely
alleged deceitful collegialism system (Degli
Esposti et al., 2010), the attempt to manage a
complex demographic phase in academic
personnel structure is probably the most
urgent issue at stake. This institution is
expected to provide contributions in the next
few years as they move toward a general
process of downsizing. In fact, strictly
connected with the theme of habilitation, there
is the question of organizational chart points
or punti organico. This system is a
comprehensive system of data to let any
public university have, annually, and is a
calculation of points to be spent for new
recruitment (both scholars, managerial, and
clerical personnel) in accordance with
retirements, performing indicators, and
financial conditions.
Actually the punti organico system is the
rationale that should assure the financial
sustainability and the differentiation of
universities in a more performance-led
criterion. Therefore, habilitation within this
framework should just be used as a filter to
have fewer pretenders and to filter out those
who do not reach some minimum level of
requirements. Formally, habilitation does not
play any further role. In fact, the total amount
of habilitated per year should affect the same
chances of recruitment. An examination of
previous data revealed that for the year 2013,
the Ministry has granted 445.5 points to all
Italian public universities, which equals to
about 445 positions as full professors or 636.3
positions as associate professors or a
2 Actually the main quest to have the best person in
an opened position recalls the mertonian assumption
of universalism whose one of the most empirical
works were made by Scott Long (Long and Fox,
1995; Long 1978).
combination of the two (full professors equals
1.0 point, associates 0.7). This is assuming
that no further administrative staff would have
been needed. Thus, this issue contains an
essential question: reproduction, extinction, or
prosperity of scientific (factions of) epistemic
communities3.
In this paper the first wave – the 2013
second wave has just ended its submission
phase – of habilitation is analyzed4 in order to
get the first empirical evidence from who did
participate in 2012 The data analysis for
reviewing this results is divided into four
sections. Beginning with data set one, a
description of the construction of the data set
is offered. In section two, technical questions
regarding the dataset and reliability of the
variables that were used are provided for the
reader. An illustration of the Italian
demographic pyramid follows in section three
in order to provide the reader an oversight of
the phenomenon of reproduction of scientific
communities. Finally, section four provides
both frequencies as well as an in depth
analyses of an exploitation of the data. In
particular, it describes a two-steps regression
test that was conducted using the primary
hypotheses of worthiness of the evaluations
by insiders and outsiders by controlling for
their respective ages as a variable. In this
section, some additional illustrations about the
rules are also provided in order to assist the
reader in better understanding if, and to what
3 As Pezzoni et al. (2012) pointed out, the list of
scientific disciplines has changed over time according
to bargaining between official representative of
scholars and the ministry. Recently a policy of
reduction of the number was pursued in order to
avoid high fragmentation. 4 At the moment of data collection, only a small
percentage of results have been released and so forth
the universe of the phenomena is not given (epistemic
communities are 179).
Marini G., Working Paper Cnr-Ceris, N° 03/2014
8
extent, the concept of worthiness can be
applied. Finally, some critical, although
preliminary and in progress, remarks about
the new institution of the habilitation are
offered in the conclusions section.
2. METHODOLOGY.
THE CONSTRUCTION OF A UNIQUE
DATASET: METHODOLOGICAL
EXPLANATIONS
Data about habilitation and its procedure,
which is basically a research and scientific
profile evaluation, are totally free and
available. Nonetheless they are designed to be
seen and not necessarily to be processed
statistically. For this reason, a track of the
construction of the dataset is provided in the
following section followed briefly by a
description of the variables used in the
analysis.
Available Data. Data about employees in
Italian universities5 including both people
who tried to get the habilitation and people
who did not are publicly available. This
information includes such demographics as
first and last names, scientific sector, faculty,
department, university affiliation, and position
(full professor, associate professors, assistant
professors (ricercatori), fixed-term assistant
professors, and statistically minor contractual
figures). However, gender (even though this
could be inferred from their respective first
names), age, or other indices of scientific
performance are not provided at the individual
level. Considering the recent reforms and
fusions, the affiliations concerning
5 Data about post-doc researchers or lecturers do not
match and do not have temporal indications. For this
reason, being all these weaker characters, they are
considered “outsiders”.
departments and faculties are problematic and
were, therefore, not considered in this study.
Conversely, in a different repository
belonging to the Ministry, curriculum vitae of
people who presented themselves for the
habilitation in 2012 is available for review.
These two sets of data, however, are not
equivocal. For example, one segment of a
university may have many employees who did
not apply (for instance, full professors do not
have many reasons to apply) whereas many
other people may have applied without
necessarily being affiliated with any Italian
university, at least as an employee. To this
latter subset, a large portion of these
individuals are expected to be post-doctoral
students, fixed-termed professors, employees,
or other non-standard personnel working in
other places (centers with main missions in
research and development such as the
National Research Council and the like).
Considering that the fiscal code is not
provided for both employees and candidates
to habilitation, no simple merging of these
two datasets could therefore be conducted. To
this regard, names and surnames were pasted
into unique strings using a final position
number ranging from 1 to 14 that served to
identify the scientific area of the person.
Assuming that any employee could not stay in
two areas and that few people may have
applied for different sectors belonging to
different areas (actually it happened, but
numbers are very little), this simple device
actually reduced the amount of homonyms.
The actual number of identified homonyms
was reduced to138 within both of the two
large datasets (around 100.000 rows). These
repeating names have therefore, been,
disambiguated manually.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
9
Also at this point, any matching identifiers
have been considered valid, thus assuming
that errors are almost absent. Moreover, non-
matching names from habilitation dataset
have also been considered external or outsider
(variables from now on are labeled as who).
The scientific standing of the latter cannot be
assumed to be necessarily inferior of those
who are insiders (see Table 5 for more
information). A final consideration is the
indispensable social capital6 of the externals
that can be assumed to be inferior due to,
either separately or conjointly, and included
both age and distance from the academic
communities.
For clarity, a description of the construction
of the matrix is now provided7. A dataset
composed of employees in academia, [A] was
matched to a dataset made up of
“applications” X “attributes concerning the
single application” called [B]. [B] has more
observations than the actual number of
persons involved, since a person has the right
to apply to more sectors and/or levels
(associate and full).
At this point a merging [A] ∪ [B], called
[C], was done: this step has 93040
observations deriving from 58060 ([B])
applications and 57518 employees ([A]).
Total is less than the sum of rows because
6 Pezzoni et al. (2012) refers and stresses the concept
of credit and operationalize it in a different research
design with data concerning for instance the scientific
collaborations. Here I don’t control for this variable
but it is assumed that habilitation ought not to given
or denied according to political capital. In fact
habilitation is not a recruitment procedure but only a
pre-selection. Moreover their design looks to be more
viable for hard sciences and much less for humanities
and social sciences, while my aim is to compare
scientific disciplines. 7 The appendix show sources, list of variables used
for this work and the diagram here designed.
candidates who are employees in universities
match. [C] has three main subjects: people in
search of habilitation who are not employees
called [C1] ([B – A]) made up of applications
as observations; not-applying people who are
employees [C2] made up of individuals as
observation ([A-B]); applying people who are
employees [C3] ([A ∩ B]), made up of
applications as observation.
This variable (called “who”) has then been
considered relevant to understand to what
extent the habilitation committees were
impartial in affording the habilitation
according to this affiliation status. Thus a
consequential job was made in order to have a
more homogenized dataset.
A matrix [D] was elaborated having only
persons as observations, making collapse rows
that had the name text in In order to store all
of the previous information obtained, 358
dummies (two for each scientific sector, that
is, for each couple for the two levels of
habilitations) were computed.
So the sum of the rows per person provided
the number of applications made by each
individual.
Once the results by candidates had been
published, other processes were completed in
order to insert the productivity variables. This
step resulted in [F] providing a row for the
analyzed privileges by scientific sector.
Available Variables. For this study, the
results of habilitation are not just provided as
a list of dummy tables (e.g. has/has not
attained the habilitation), but instead included
three indicators of scientific productivity that
are similar to other studies (Ginther & Kahn,
2004). These variables added from the data
concerning the evaluations of applications
were separated between the hard sciences
(areas from 1 to 9) and the soft sciences
Marini G., Working Paper Cnr-Ceris, N° 03/2014
10
(10 to 148). The hard sciences include: (a) an
H index (normalized by academic age9; (b)
normalized number10
of articles; and (c)
normalized number of citations. The soft
sciences have: (a) normalized number of
books; (b) normalized number of chapters in
books and articles; and (c) normalized number
of articles published in top ranked reviews11
.
In cases were curriculum vitaes were stored as
pdf files, they were downloaded one by one
and may have offered further information
such as date of birth, number of projects led or
coordinated, professorships held, or other
salient information about possible further
academic titles (fellowships, editorial
memberships, etc.).
Nonetheless, this last source of information,
curriculum vitaes, was used only to ascertain
the ages (for a total of around 7000
observations) of the individuals, due to the
difficulty obtaining this information using
existing data sources. Regarding the reliability
and validity of the data, the three indicators
used in this study were assumed to have an
8 Exceptions, not given here due to not-yet-resealed
data concerning the following sectors, are: 8C1, 8E1,
8E2 and 8F1 that don’t have bibliometric indices
(they are Design and Architecture disciplines); 11E
disciplines have bibliometrics (Psychological
sectors). 9 For an in depth analysis of H index by age see
Mannella and Rossi (2013). 10
Normalizations are basically referred to a measure
of personal contribution to an output when the latter
are signed by more persons. 11
Having been debated recently about the misuse of
citation indices in social sciences, evaluation of
research and habilitation in social sciences opted to
use a general criteria of peer review. In this case, an
attempt was made to discriminate general
productivity – alleged to be in most of the cases of
modest value, parochial and poor in originality and
innovation – from a bargained list per sector of top-
ranked reviews.
intrinsic validity of productivity and quality of
research, while noting that they cannot tell
about networks of collaborations (social
capital for Pezzoni et al. 2012) nor
specialization in topics (Leahey, Keith, &
Crockett, 2010). The H indexes and articles in
top-ranked journals were assumed to measure
quite well the quality of scientific production,
while the others can be assumed as a good
proxy of productivity in terms of quantity of
outputs, with uncertainty about its’ quality.
3. THE EMPLOYEES PYRAMID OF
ITALIAN UNIVERSITIES
The possibilities for new concorsi
(competitions for permanent positions) are
strictly associated with the current amount of
professors now employed. Projections
published by the Minister foster additional
data about how many people are supposed to
leave for retirement through 2016. This
projection is based on the age of each person,
since retirement is basically mandatory in
higher education by age 70, regardless of the
years accrued in service.
An exception to this rule can be a
professor’s claim for a two year extension in
order to continue their work with full titles in
their respective chairs. Lately, these
procedures, however, are becoming
increasingly difficult since applications for
extensions are no longer a formal quest and
whose decisions are no longer considered
quite certainly affirmative. In fact, the last
general law (240/2010) tried to reduce these
phenomena even though no data was available
about the impact of such a change. Moreover,
it is even less certain the number of people
who are applying for or who will apply in the
future as a condition of pre-retirement.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
11
Table 1 Indicators of Demographics by Some Scientific Sectors and Results of Habilitation
(Selected Scientific Sectors are Among Those that Were Published First)
total #
of expected
vacancies
ratio of full prof.s
over associate prof.s
ratio of full prof.s over
associates+ (permanent)
assistant prof.s
ratio of associates prof.s
over (permanent)
assistant prof.s
ratio of attained
habilitations in 1° level
ratio of attained
habilitations in 2° level
habilitations attained
(both levels) / expected
vacancies (# persons)
# of total attained
habilitations
(both levels)
01/A4 32 1.0100 0,5260 1,0870 43,1% 39,4% 478,1% 153
02/B1 54 0.6923 0,3172 0,8455 73,0% 78,1% 1042,6% 563
02/B2 24 0.5400 0,2983 1,2346 71,2% 69,7% 1370,8% 329
03/C1 53 0.8909 0,3443 0,6298 51,0% 51,7% 300,0% 159
06/D3 49 0.8557 0,2660 0,4512 31,1% 41,5% 336,7% 165
06/E1 38 1.0256 0,3980 0,6341 29,2% 37,3% 263,2% 100
07/H1 10 0.5397 0,2482 0,8514 65,9% 52,5% 580,0% 58
07/H3 8 0.8478 0,2977 0,5412 69,7% 76,6% 1025,0% 82
07/H5 6 0.7872 0,3058 0,6351 62,1% 70,0% 883,3% 53
08/A1 28 1.2273 0,5000 0,6875 34,9% 45,7% 303,6% 85
08/B3 33 0.9425 0,3923 0,7131 50,0% 44,2% 218,2% 72
09/H1 31 0.8920 0,3870 0,7662 36,9% 42,6% 877,4% 272
11/A1 27 1.2083 0,4567 0,6076 38,8% 34,9% 292,6% 79
11/A3 80 1.2333 0,5000 0,6818 34,5% 40,7% 266,3% 213
11/A4 31 0.7143 0,3636 1,0370 45,0% 28,6% 458,1% 142
11/C2 26 1.5952 0,7791 0,9545 37,6% 40,3% 442,3% 115
11/C4 20 0.8983 0,3786 0,7284 30,4% 44,2% 605,0% 121
12/B1 34 1.6289 0,5745 0,5449 56,1% 31,2% 264,7% 90
13/A5 8 2.2143 0,8611 0,6364 63,4% 56,5% 812,5% 65
14/C1 67 0.8392 0,3125 0,5934 37,8% 29,0% 267,2% 179
Total of available
data 659 1.0291 0,4253 0,7430 48,1% 47,7% 469,7% 3095
Total academic
population 54930 0.8996 0,3594 0,6653 n.a. n.a. n.a. n.a.
Source: own elaboration on MIUR data. Total academic population refers to all scientific sector (179); Total only to shown sectors
Note 1: for assistant professors only permanent ones are given because the fixed-term assistant professors are still a decisive minority
Marini G., Working Paper Cnr-Ceris, N° 03/2014
12
Even though this phenomenon was definitely
not traditionally common, some incidents may
continue to occur and may have an impact
even if the smallest level of epistemic
community is taken into account (few
numbers count a lot in lower layers).
Despite all of these caveats, a gross and
rounded up number of new positions that can
be used in the next few years – more or less
during the same years the habilitation will be
valid (4 years since the day of publication of
results) – can be computed. All estimations
here have to assume that, however, new
habilitations will be given annually thus
creating a chained longitudinal series of
people who will have title to apply for
associate and full professorships. Hence,
larger numbers by sector in the projected year
of 2016 (here given in Table 1) equals a
stronger struggle for positions.
As indicated in Table 1, the total of
employees in Italian universities at the end of
2012 was about 55,000 full time units. The
pyramid is quite critical as the ratios among
levels may indicate. In fact, for any associate
professor there were almost 0.9 full
professors, about the same amount as there are
currently. For assistant professors, who must
all get the habilitation to improve their
careers; there are less than three times their
ranks compared to full professors (0.36). The
associate professors level, though, is about
two-thirds the level of assistant professors.
Strong differences among scientific sectors,
the actual sub-area communities of peers who
are claimed to manage themselves through the
new rule of habilitation, are therefore very
clear. In some cases, there are more full
professors than there are associates (ratio over
1). For instance, in the epistemic community
11C2 (Logic, History, and Philosophy of
Science), the number of full professors is very
near to the total of the other two layers
summed together (0.78). Also, the ratio of
associates and assistant professors does not
reveal a scattered diversification in the
scientific sectors that have been previously
analyzed (as a range: max 1.23; min 0.45).
Percentages of attained habilitation over the
total number of applications can vary from
less than 33% to 78%. This is a remarkable
difference whose explanation is not easy to
identify and whose reasoning may lie on the
particular disciplines peculiarities or uses and
interpretations of the new institution of
habilitation.
An interesting index is the ratio of
habilitations given and the vacancies that each
scientific sector will experience. To this
regard, it can be seen that in many cases the
habilitation of a sole wave are two or three
times the vacancies of several years (for
instance 3/C1). In other cases, the
habilitations are more than 10 times the
number of expected ceased positions (for
instance in 2/B2). Considering the only real
number possible12
, the overall recruitment
points for all university and sectors for 2013
could be 445 points that can be used for more
than 3,000 habilitations, thus the sectors here
are only 20 out of 179. As a result, it is clear
that even though there are huge differences
between sectors, the winners of habilitations
have only overcome a small hurdle; however,
they have not accomplished their final
endeavor. At the same time, the habilitation
here seems to have become a system that
signals who can enter (or improve their
12
http://attiministeriali.miur.it/media/227960/tabella
punti_organico_2013.pdf.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
13
career) and more importantly, who cannot13
.
This seems especially true for those
individuals who are not awarded with
habilitation and thus have the minimum
requisites to access the habilitation (e.g. you
are not that bad, but we don’t want you
neither as a competitor). This fact is relevant
since the next competitions will be hosted
locally and will not be on a national level.
This procedure might also serve to reintroduce
by the by default the antiquated practice of
parochialism, with the innovation that a
preselected list of competitors had already
been excluded.
The most critical point here is the
impossibility to renew with the same pace of
retirements these scientific communities and
thus losing the opportunity of having a stable
amount of young (as the fixed term assistant
professorships are not many until now, the
average age of new habilitated and
consequently new associate professors will be
probably not so young, and most probably
already mature) cohorts. It must also be
mentioned that the average age of most Italian
scholars is already seriously high (medians for
full, associate, and assistant professors are
respectively: 60, 52, 44 with means of 58.9;
13
People who tried to get the habilitation and failed
are proscribed to participate to the same sector for 2
consecutive years. Future attempts can so forth be
done even though minimum requisites are reached
without attaining the habilitation. But just for that, a
message like this can be interpreted by the loser: “in
the better of cases, mind that you are backward in the
queue”. As recruitment in higher education system
will be shortly describe, even this more optimistic
interpretation equals to a de facto exclusion, unless
priorities in the queue (one’s credit over other who
could have won now, and hadn’t the chance to get a
post meanwhile) wouldn’t change. In a logic of
factions within a sector, this interpretation is possible.
52.9; 45.4; Rossi 2012a). Thus, all this
process of economic efficiency – the expected
saving is not relevant since the pyramid has a
bulge of people near mandatory retirement
age and since traditionally the recruitment in
Italy has witnessed long pauses coupled with
large waves of open positions (Lissoni et al.,
2011) – probably will intensify a problem
which is connected to brain drain and
capability of an epistemic community to
generate new internationally cutting edge
knowledge. Even this latter implication is not
included as a further analysis, it is essential
since average age of human resources in
research pertains the capability of epistemic
communities to continue to exist (Pezzoni et
al. 2012).
4. DATA ANALYSES: WERE
COMMITTEES A SOURCE OF
MERITOCRACY?
A First Overview. As previously described,
the construction of the dataset required the
researchers to divide the applicants amongst
those who were already employed by the
university from those who were hoping to
obtain tenure.
With this in mind, data in Table 2 helps
reveal that the largest portion of employees in
universities applied, especially if it is taken
into account that full professors do not have a
desire to compete for this award (percentage
of applying employees in Italian universities
is 78.98%).
The impressive amount of applications
among employees in universities and from
other contexts (or with academic but more
informal or weaker affiliation), possibly
suggesting that the first wave might have been
interpreted as an occasion to have a try.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
14
Table 2 Description of the Universe: Employees in HEIs and Applicants
a.v. %
a) Employees in HEIs not applying 34978 44.67
Full professors 13239 16.91
b) Applying employees in HEIs 22538 28.78
c) Applying employees in HEIs (without full professors) 22538 78.98
d) Applying not employees in HEIs 20794 26.55
Total (a+b+d) 78310 100.00
Source: own elaboration on MIUR data.
The persons who do not have a position as
an employee in a university were revealed to
have more of a propensity to apply to more
than one sector/level.
This is a quite reasonable occurrence in that
their career paths may have not been
solidified or their profile may not have
perfectly matched those of the Italian
academic system. As a result, more attempts
and paths could have been chosen.
Data in Table 3 illustrates that employees
applying for a professorship in most cases
tried only one sector with only three
individuals applying to six different
sectors/levels in the first level, and only one
person applying in seven sectors among the
second levels. In contrast, among the
outsiders the tail is much longer and the
percentages of those who limited themselves
to only one submission is restricted when
compared to their insider competitors.
For example, 100 of the persons employed
in Italian academia who participated in
habilitation, 87.4 and 86.8 respectively, made
only one application respectively for first and
second level (Table 3).
The same percentages for people who are
not inside universities as employees were 81.2
and 79.5, respectively. This can be considered
a much more frequent propensity by external
submitters to try more options in a tentative
way, even though this does not imply weaker
scientific standings.
Understanding this implication is critical for
the next step of the analysis in that the
variable under discussion (insiders vs.
outsiders) is introduced with the indicators of
scientific production in order to test if, having
same scientific productions, to be a part or not
to be a part of academia is a good predictor of
the outcome in question: attainment or no
attainment of the habilitation.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
15
Table 3 Number of Applications Submitted by Typology of Actors (Who Variable14
)
and by Level of Application (Full Professors and Associate Professor)
first level second level
# of application insiders outsiders Total insiders outsiders Total
1 8.882 3.804 12.686 12.297 14.683 26.980
87.42 81.26 85.48 86.82 79.52 82.69
2 1.022 649 1.671 1.465 2.775 4.24
10.06 13.86 11.26 10.34 15.03 12.99
3 196 159 355 295 701 996
1.93 3.4 2.39 2.08 3.8 3.05
4 38 35 73 70 202 272
0.37 0.75 0.49 0.49 1.09 0.83
5 19 18 37 32 62 94
0.19 0.38 0.25 0.23 0.34 0.29
6 3 7 10 4 26 30
0.03 0.15 0.07 0.03 0.14 0.09
7 4 4 1 6 7
0.09 0.03 0.01 0.03 0.02
8 1 1 5 5
0.02 0.01 0.03 0.02
9 1 1 1 1
0.02 0.01 0.01 0.00
11 1 1 2 2
0.02 0.01 0.01 0.01
15 2 2 1 1
0.04 0.01 0.01 0.00
18
1 1
0.01 0.00
Total 22,538 20,794 43,332 22,538 20,794 43,332
100.00 100.00 100.00 100.00 100.00 100.00
Source: own elaboration on MIUR data.
14
For simplicity, labels of “who” variable are sometimes different as the meaning of this same variable can be
of being “insiders” of “outsiders” of the system, or more technically but less shortly, employees and not-
employees in universities as researchers.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
16
An examination of the data affirms that
having the same standing of curriculum vitae,
both an insider and an outsiders should have
the same chances of being accredited as
habilitated in that a recruiting agent – both an
institution or a unit like a department – cannot
claim in this step a question of compatibility
in terms of research interests, previous
partnerships, nor credits meant as relations
between the candidates and the seniors.
This is defined as political capital according
to Pezzoni et al. (2012). In fact, the habilitated
will not necessarily work with the members of
the commission even though the committees
can shape and influence the features of their
epistemic community by giving or not giving
this title. It is possible, then, to hypothesize
that judgments based strictly only of scientific
production indicators may not happen due to
an enclosure mechanism that will favor those
who are already inside the system at the
expense of those who have not yet entered the
system.
These can mean then for those candidates
with a strong standing in terms of scientific
outputs, the result could even result in a
change of employment from a non-Italian-
academic entity to the Italian university
system itself.
As previously stated, this exclusion
mechanism can strengthen the informal
institution of the queues of pupils listed by
affiliation with their inbreeding processes
(Pezzoni et al. 2012). Even so, habilitation
could overcome this mechanism, replacing
this practice with mere worthiness based on
production or rethinking all this in a
reconfiguration of the discipline based on the
power of peers.
This process cannot be tested in this study,
however it is presented as a formal reminder
to those people who already have a better
position and/or are already with one step
inside the system that they would somehow a
receive a reasonable chance even though it is
not coherent with the juridical system and the
specific norms ruling the institution of
habilitation. In fact, habilitation sees indices
of individual performance as the main criteria
of evaluation, yet the three indices it aims to
use filters out people below some of the
indicated good threshold while at the same
time it is expected to filter in people over that
same good threshold.
Even though peer reviews of selected
scientific outputs can affect the evaluation of
candidates having the same numbers of
outputs, and even though further credits in
one’s curriculum vitae may be relevant, some
descriptive statistics by insiders and outsiders
are now indispensable.
Equally important, the outsiders have higher
standard deviations (with the exception of
number of articles in top-ranked journals for
science and humanities candidates: SD =
4.023 for insiders and SD = 3.018 for
outsiders) in all indicators of productivity,
both in the hard sciences as well as in the
social sciences and humanities (see Table 4).
There is, however, a clear difference between
candidates in scientific disciplines belonging
to hard sciences (measured by bibliometric
indicators) and the others.
For instance, the outsiders in hard sciences
have – just as a mean – better indicators,
while in social sciences and humanities the
evidence reveals the opposite: insiders have
higher productivity in all three indicators.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
17
Table 4 Descriptive Statistics of Productivity by Typology of Candidates
(Insiders and Outsiders)
Insiders Outsiders
obs. mean S.D. obs. mean S.D.
articles 2718 36.926 31.706 2591 37.203 34.2455
citations 2717 49.998 66.749 2591 55.659 77.8591
H index 2717 10.314 5.9263 2591 10.593 6.2509
books 1027 3.051 2.4370 1320 2.831 3.3634
chapters & articles 1027 21.339 16.1746 1320 17.051 25.8337
articles in top ranked journals 1027 2.850 4.0235 1320 1.698 3.0177
Source: own elaboration on MIUR data
For this reason, the use of the variable who
is quite problematic since the outsiders are in
any case, even after splitting hard and soft
sciences, a heterogeneous group whose
composition cannot be analyzed easily. Both
Figure 1 and Table 5 reveal the distribution of
ages by insiders and outsiders. Overall,
outsiders are younger by 4 years and the first
and last quartiles are also both younger by 3
years. The outsiders’ distribution has an even
higher kurtosis and a slightly higher skewness
level.
Literature about age and productivity is
nonetheless an old preoccupation, especially
in the United States. Clemente (1973)
discussed findings dating back to the 1940s
and 1950s affirming that early publications
would be a good proxy of potential for a one’s
career thus early publication is a sign of high
career research outputs. More recently,
Bozeman, Dietz, and Gaughan (2001)
reported that by focusing only in hard
sciences and technology laden professions,
nowadays post-doctoral students (which
would be here outsiders) are not the best
potential candidates for research and
development since career trajectories bring
young adults to have their best laboratory
experiences, even outside academia. To this
regard, this differentiation between who is in
and who is out academia may have some
relevance even though the United States and
Italy differ in many respects.
Levin and Stephan (1991) reported that
productivity cannot depend on age due to
various scientific sectors and, especially,
having different decades and different paces
of productivity over a scientists' life course.
The pace of accrued outputs is not constant
and empirical evidence brought these
researchers to highlight the relevant role of
investments in research and development.
Even though the numbers of scientific outputs
were eligible only if no more than 10 years
old (since 2003 until 2012 is included), the
numbers of publications may be affected by
the author’s younger age, especially for
younger scholars trying to get the habilitation
for associate positions15
.
15
Some Committees displayed information about
personal years spent as research active, considering
the starting age (first publication) minus official
maternal/parternal leaves. Having no gender purposes
here and making the regressions by scientific
disciplines, this aspect was omitted.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
18
Table 5 Descriptive Statistics of Age of Applicants by Typology of Candidates
Insiders Outsiders
Obs. 3427 3582
25% 40 37
Median 46 42
75% 52 49
Mean 46.604 43.280
S.D. 8.0047 7.9062
Skewness 0.39138 0.51002
Kurtosis 2.39577 2.56251
Source: own elaboration on MIUR data
For these reasons, and even considering the
reduced availability of information for this
dataset, age was use to check (through a
Heckman two steps treat) the insider/outsider
variable, with more recent findings reporting
concerns between the links of age, cohorts,
and periods tested in science productivity
(Hall, Mairesse, & Turner, 2005).
Source: own elaboration from MIUR dataset
Figure 1. Distribution of candidates by age and insiders (labeled 1) and outsiders (labeled 0)
Marini G., Working Paper Cnr-Ceris, N° 03/2014
19
What really matter to get the habilitation?
Tables 6a and 6b provide the final stage of
this analysis for this study. The identities of
applicants, labeled as who, is inserted into a
full model of several regressions with the
three indicators along with the who variable
that was previously treated for age. Indicators
of productivity are used as they appear from
the personal folders of the candidates:
continuous variables measuring the
quantitative/qualitative production of
scientific production.
Committees had to check if the medians16
were reached or not in order to decide. This
resulted in a person being judged habilitated
only if two of the three indicators had been
reached or if performance indicators
surpassed the medians among the sector in
question.
These indicators included those candidates
whose production over the last 10 years does
not overtake neither one median of his/her
sector or should be awarded with habilitation
only if other criteria (always publicly
expressed by the committees, but they refer to
a peer review job sustained by the further
credits detailed in the candidates curriculum
vitae) did not lead to a different inducement.
Hence, the hypothesis concerning scientific
production as being not the only predictor of
16
“Medians” turned recently to become ynonimous
of “thresholds” in Italian debate. The national
Agency for evaluation and accreditation published
these values by scientific sector before the release of
results: http://www.anvur.org/index.php?option=com
_content&view=article&id=253:asn-indicatori-e-relat
ive-mediane-it&catid=13:sitoit&Itemid=314&lang=it
These values as said are subjected to peer review
made by the members of commissions and could be
contracticted.
attainment of habilitation does indeed deserve
a more in depth analysis17
.
However, I would caution that an apparent
identical hypothesis could be as follows:
people who are already in academia as
employees may be successful if just minimum
threshold are covered but are not so strong in
relative terms when compared with other
scholars.
This second version differs from the original
hypotheses in that it excludes who is,
apparently from indicators, good but is not
recognized as eligible to compete to become
part of the community.
An additional consideration thing is though
must be given to those eligible people who are
barely good, but are recognized as already
have obtained some position to enter as a peer
in the academic community.
Last, the theoretical assumption here is that
the Italian system, even though deeply
reformed by the last general Law, remains
regarding the career ladder a regular
employee track and did not opt to change into
a contract track or a tenure track (Enders,
2001).
To this regard, tables 6a and 6b helps to
identify which committees discriminated or
did not discriminate against candidates by
their current positions.
17
At the same time, an apparent identical hypothesis
could be as follows: people who are already in
academia as employees may be successful if just
minimum threshold are covered but are not so strong
in relative terms if compared with other scholars.
This second version differs because one thing is to
exclude who is, apparently from indicators, good but
is not recognized as eligible to compete to become
part of the community; other thing is to let consider
eligible people who are barely good but are
recognized as already somewhere in the queue to
enter as a peer the community.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
20
Table 6a – Logistic regressions for habilitation attained: models with scientific production
indicators plus status of “insiders/outsiders” dummy variable treated by Heckman
two steps by age (bibliometric scientific sectors)
obs. model Articles Citations H_index Insiders/outsiders age
label of sector
coeff. s.e. coeff. s.e. coeff. s.e. coeff. s.e. coeff. s.e.
1_A4_1 Mathematical
Physics
144 0.9389 0.000 (0.002) -0.002 (0.015) 0.005 (0.015) -0.176 (0.740) 0.019 (0.013)
1_A4_2 231 0.0619 0.003 (0.003) -0.006*** (0.002) 0.028 (0.019) -0.958 (0.749) 0.028** (0.013)
2_B1_1 Applied Physics
of matter
230 0.0000 0.002* (0.001) -0.000 (0.004) 0.022** (0.009) -0.297 (0.239) 0.052*** (0.013)
2_B1_2 505 0.0000 0.001 (0.001) -0.001** (0.001) 0.046*** (0.008) -0.027 (0.156) 0.050*** (0.010)
2_B2_1 Theoretical
Physics of
matter
139 0.0000 0.002** (0.001) 0.000 (0.001) 0.025** (0.242) -0.326 (0.242) 0.056*** (0.016)
2_B2_2 330 0.0000 0.003 (0.002) 0.001 (0.007) 0.036*** (0.009) -0.103 (0.303) 0.036*** (0.012)
3_C1_1 Organic
Chemistry
104 0.0000 0.004* (0.002) -0.000 (0.001) 0.036* (0.019) -0.407 (0.726) 0.040 (0.025)
3_C1_2 205 0.0000 -0.001 (0.002) -0.001 (0.001) 0.072*** (0.015) -0.273 (0.466) 0.030* (0.015)
6_D3_1 Blood diseases,
Oncology and
Rheumatology
121 0.0000 0.002** (0.001) -0.001 (0.001) 0.034*** (0.009) 0.164 (0.275) 0.049*** (0.018)
6_D3_2 306 0.0286 0.001 (0.002) 0.000 (0.001) 0.035 (0.023) -4.240 (13.718) 0.003 (0.010)
6_E1_1 Cardiovascular
and Thoracic
Surgery
96 0.1610 -0.001 (0.002) 0.000 (0.002) 0.028 (0.025) -1.031 (1.280) 0.023 (0.018)
6_E1_2 193 0.9821 0.003 (0.019) -0.002 (0.021) 0.052 (0.201) -16.743 (144.037) 0.002 (0.013)
7_H1_1 Anatomy and
Physiological
Veterinary
41 0.0000 0.002 (0.006) -0.006* (0.004) 0.109*** (0.022) -0.153 (0.846) 0.043 (0.048)
7_H1_2 59 0.0000 0.013*** (0.006) -0.011* (0.006) 0.133*** (0.034) -0.645 (0.497) 0.053 (0.030)
7_H3_1 Infectious and
Parasitic
Diseases of
Animals
33 0.6382 -0.001 (0.006) -0.001 (0.004) 0.054 (0.054) 2.765 (4.567) -0.022 (0.037)
7_H3_2 77 0.0088 0.002 (0.003) 0.000 (0.003) 0.039* (0.022) 0.496 (1.039) 0.018
(0.020)
7_H5_1 Clinic Surgery
and Animal
Obstetrical
29 0.0828 0.015 (0.009) -0.006 (0.010) 0.060 (0.071) 0.056 (1.511) 0.056 (0.069)
7_H5_2 50 0.1337 0.002 (0.015) -0.025 (0.028) 0.195* (0.105) 2.403 (2.893) -0.024 (0.032)
8_A1_1 Hydrology,
Hydraulics,
Hydraulic and
Nautical
Buildings
63 0.0000 0.015** (0.007) 0.004 (0.004) -0.009 (0.031) 1.138*** (0.362) 0.115*** (0.039)
8_A1_2 138 0.0000 0.006 (0.007) -0.005 (0.005) 0.082*** (0.028) -0.080 (0.198) 0.103*** (-0.022)
8_B3_1 Building
techniques
60 0.7700 0.011 (0.029) 0.003 (0.029) 0.049 (0.127) -5.131 (22.697) -0.006 (0.030)
8_B3_2 94 0.0011 -0.007 (0.007) 0.00 (0.007) 0.097*** (0.032) 1.242 (1.453) 0.018 (-0.021)
9_H1_1 Elaboration of
Information
Systems
260 0.0011 -0.006* (0.004) -0.000 (0.002) 0.064*** (0.023) 1.863 (3.677) 0.007 (-0.014)
9_H1_2 412 0.0000 -0.006** (0.003) -0.003** (0.001) 0.083*** (0.013) -0.227 (0.631) 0.019* (0.012)
Source: own elaboration on MIUR data
* p<0.10
** p<0.05
*** p<0.01
Marini G., Working Paper Cnr-Ceris, N° 03/2014
21
Table 6b. Logistic Regressions for Habilitation Attained: Models with Scientific Production
Indicators Plus Status of Insiders/Outsiders Dummy Variable
(not-Bibliometric epistemic communities)
# books # articles or chapters # articles in top-
ranked journals insiders/outsiders age
obs. model coeff. s.e. coeff. s.e. coeff. s.e. coeff. s.e. coeff. s.e.
11_A1_1 Medieval History
49 0.9769 0.004 (0.024) -0.001 (0.005) -0.012 (0.027) 0.473 (0.639) 0.047 (0.032)
11_A1_2 172 0.0000 0.010 (0.014) 0.010*** (0.003) 0.053*** (0.017) 1.396 (2.302) 0.009 (0.015)
11_A3_1 Contemporary History
116 0.0000 -0.002 (0.011) 0.006*** (0.002) 0.073*** (0.020) 0.559* (0.330) 0.037** (0.016)
11_A3_2 425 0.0000 0.016** (0.008) 0.005*** (0.002) 0.119*** (0.014) 0.002 (0.263) 0.028*** (0.008)
11_A4_1 Book and Documents
Sciences; Religion
Sciences
111 0.0006 -0.007 (0.015) 0.000 (0.002) 0.072*** 0.003 -2.498 0.000 0.052*** (0.017)
11_A4_2 322 0.0000 0.005 (0.009) 0.008*** (0.002) 0.044*** (0.011) 1.074 (0.808) 0.016 (0.016)
11_C2_1 Logic, History and
Philosophy of Science
85 0.0001 0.002 (0.014) -0.001 (0.001) 0.047*** (0.011) 0.633 (0.583) 0.031* (0.018)
11_C2_2 205 0.0000 0.011 (0.014) 0.005** (0.002) 0.082*** (0.012) 1.386 (1.087) 0.017 (0.012)
11_C4_1 Aesthetics, Languages
Philosophy
69 0.0008 0.010 (0.013) -0.002 (0.002) 0.071*** (0.016) 0.241 (0.415) 0.045** (0.023)
11_C4_2 226 0.0002 0.020** (0.010) 0.005** (0.002) 0.033*** (0.012) 1.020 (0.889) 0.013 (0.011)
12_B1_1 Commercial and
Nautical Law
57 0.1085 0.036 (0.051) 0.005 (0.007) 0.005 (0.015) 2.183* (1.240) -0.036 (0.026)
12_B1_2 186 0.3906 -0.010 (0.017) -0.002 (0.004) 0.013 (0.011) 1.786 (1.303) -0.020 (0.016)
13_A5_1 Econometrics
37 0.9795 -0.074 (0.425) -0.009 (0.038) 0.053 (0.102) -7.097 (70.301) 0.003 (0.026)
13_A5_2 60 0.1748 -0.048 (0.075) -0.000 (0.008) 0.041* (0.022) 0.348 (1.576) 0.014 (0.023)
14_C1_1 Genera, Juridical and
Political Sociology
117 0.0000 0.001 (0.014) 0.007*** (0.002) 0.028*** (0.008) 0.027 (0.354) 0.052** (0.021)
14_C1_2 356 0.0000 -0.011 (0.008) 0.000 (0.002) 0.034*** (0.006) 0.351 (0.811) 0.008 (0.010)
Source: own elaboration on MIUR data
* p<0.10
** p<0.05
*** p<0.01
All the three indicators of productivity as
well as the who variable were used regardless
of the extent to which these indicators were
above or under the medians for each
candidate. The dummy variables regarding
being insiders or outsiders as previously
explained is part of the model with the four
predictors. Tables 6 are split to differentiate
the hard sciences from the soft sciences.
Committees under investigations are 40: 20
scientific sectors by two levels. Full
professors are identified as (_1) and associates
as (_2). Additionally, eight of the sectors are
in social sciences and humanities with the
remaining number of sectors belonging to the
hard sciences. As identified in the tables, 15
of the 24 models in the hard sciences have a
statistically significant difference whereas
nine models do not show a statistically
significant difference. In 11 committees, there
is a statistically significant difference
compared to the other 5 committees that did
not reveal statistically significant difference.
The non-bibliometric sectors revealed similar
results.
In the hard sciences, 14 committees out of
24 have the normalized H index of Hirsch as a
good predictor of getting the habilitation (as
Marini G., Working Paper Cnr-Ceris, N° 03/2014
22
expected, the association is always the higher
the index, the higher the probability to get the
habilitation): 10 times at p <0.01 level. Thus,
seven times the number of articles can predict
the outputs of attainment of habilitation. In
two of these cases, the association is negative
(Elaboration of Information Systems, both
levels). In only five times did the number of
gross citations help predict the phenomena,
and in four cases the coefficient was negative.
The only positive coefficient that was
significant was at the 0.01 level
(Mathematical Physics, associate level). In
only one case, (Hydrology, Hydraulics,
Hydraulic and Nautical Buildings, full
professor level), the insider/outsider variable
had a significant and strong coefficient, with
the advantage of insiders.
Overall, conjointly 26 times did an indicator
which is formally part of the decision making
predict the verdict of habilitations. Also, only
in 7 committees predictors were not found.
Generally, the H index looks to be, in most of
the cases, a stable and affordable predictor for
getting the habilitation.
In the social sciences (Table 6b), only three
committees are left without predictors
(Econometrics, first level; Commercial and
Nautical Law, second level; Medieval
History, first level). Among those with sound
predictors, 12 times the high number of
articles published in top-ranked journals is a
predictor for the attainment of the habilitation.
Furthermore, seven times the number of other
articles and chapters in books is associated
positively with this outcome. In only two
committees were the numbers of books
somehow relevant. In any case, even in social
sciences and humanities, that cannot have
bibliometric indicators, an analogous
hierarchy between the three indicators was
noted: articles in top-ranked journals (H index
for hard sciences) are better than articles and
chapters in books (number of citations in hard
sciences), though they subsequently are better
than the number of books (number of articles).
In both the hard sciences and not-bibliometric
sector the insider/outsider does not play any
role, expect for the few sectors that should be
analyzed in more detail.
5. CONCLUSIONS AND
DISCUSSIONS ABOUT
HABILITATION
The Italian academic system in the
recruitment issue was described as a "multiple
processes of negotiations" (Nelken, 2009,
p.?), even though recent evidence about
institutional reforms, for instance the role of
evaluation, shows that recruitment is changing
into a differently bargained process (Reale &
Marini, 2013). This new relation is based on a
maussian gift (Fassari, 2009) or the political
power (Pezzoni et al. 2012) and is very
powerful, but this perspective tells only a part.
Why a big don professor should show such
generosity has long been the subject of
debate? To some extent, it’s a matter of
showing his/her power to his/her peers, as an
indirect index of strength. It is indeed even a
question of letting go of one's group at the
expense of others. It is, last but not least, a
pay-back for many services and little jobs the
junior already had given. In fact, Nelken
(2009) affirms, showing in depth knowledge
of the Italian system, especially of social
sciences in which old professors must provide
evidence to be able to give a post in order to
have someone under them do many regular
jobs (i.e. supervise undergraduates,
accomplish minor phases of research, etc.).
Marini G., Working Paper Cnr-Ceris, N° 03/2014
23
The asymmetrical relations beyond this
reciprocal obligations and its parochialism are
not even a new issues in these studies (Perotti,
2002), nor just an Italian practice, since
inbreeding is an old issue and not just an
Italian one (Combes, Linnemer, & Visser,
2008), often intertwined with the stratification
of universities and departments (McGee,
1960; Kirchmeyer, 2005). In any case, in this
traditional framework the role of the
habilitation cannot already be defined clearly.
It must be said that, despite the Italian
public opinion and even scholars’ common
debate about Italian higher education, to have
a sponsor and to have a career in universities
is nothing but a regular matter (Kyrchmeyer,
2005; Musselin, 2004; Reskin, 1979). In fact,
there are contexts as well rules to prevent bad
collegiality (Enders, 2001). Even though in
the Italian system internal market (Musselin,
2005) is formally denied, the allegations of
parochialism and nepotism by protégé haven’t
ceased so far. This first wave let emerge a
further problem, which pertains more with
legitimation of the evaluation itself. Perhaps
the words by Bourdieu (1975) can more
eloquently summarize the phenomenon:
In a highly autonomous scientific field, a
particular producer cannot expect
recognition of the value of this products
("reputation", "prestige", "authority",
"competence", etc.) from anyone except
other producers, who, being his
competitors too, are those least inclined to
grant recognition without discussion and
scrutiny. This is true de facto: only
scientists involved in the area have the
means of symbolically appropriating his
work and assessing its merits. And it is
also true de jure; the scientist who appeals
to an authority outside the field cannot fail
to incur discredit (p. 23).
It can also be said that further discussion is
mandatory. In fact, some habilitation
committees have been sued formally (appeals
by not-habilitation awarded to regional
administrative courts – namely TARs – have
already occurred and have already been
rejected) and informally (through tough
debates within some epistemic communities
or scientific disciplines, which is quite the
overlapped translation of this concept18
). An
analysis of these data cannot cause one to
shrink from some discussion of the topic.
Following the suggestion in Bourdieu’s
passage, usually, scientific communities
should react harshly to appeals to external
actors, like appealing for justice to tribunal or
exposing one’s case to punish the system in a
pillory. One explanation to these phenomena
may concern cultural aspects while others
may argue that some traits of the weak state’s
institutions play a role. At least, the possibility
of a dominant civil servants’ ethos (where
achieving to climb the career ladder is
expected to happen just by seniority) to
respect of those more typical of scholars
(based on prestige accrued through scientific
outcomes and intellectual acknowledgement)
could be envisaged. Explanations of this
phenomenon go beyond the possibilities of
this work.
18
A comprehensive and up to dated list of both these
sort of appeals and debates are listed here:
http://www.roars.it/online/documentazione-asn-e-
vqr/. In some cases, such as in sector 11A4 (Book
and Documents Sciences; Religion Sciences, here
anayzed and among those with found predictors), the
case was brought in the national Parliamant for an
interrogation concerning the procedures and criteria
adopted by the Commission.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
24
Whatever the cause, the new habilitation
doesn’t seem, especially in some not-
bibliometric sectors, to be a pacific exercise of
evaluation because the appeals and the debate
are too massive to be banished as marginal
facts. In fact, allegations and critiques already
emerged and these show that this new
institution may have weak points besides its
slowness. A specific, not statistical, example
is the case of pro-veritate19
judges in 11A4
sector. In this sector, pro-veritate was
appealed to much more often if compared to
all other sectors here taken into account.
Generally, this extra judge was practically a
mere exception20
, while in 11A4 it became
quite a systematic way to give – always to the
same person – the power to decide. This looks
to reduce the collegial negotiations to a
personal domain over a whole community. In
any case, this mentioned example only
reinforces the need to pursue further
investigations upon careers, recruitment and
evaluation in higher education.
If the system of concorsi could have been
seen as a panel game within the epistemic
community that could regulate itself any time
a concorso happened (with the capability to
give and change meta-rules in the mid long
run), habilitation looks to have an interesting
role because it is managed by few people and
just one or two persons are able to don’t
afford the habilitation (committees are
19
Pro veritate evaluations can be freely appealed by
the committees on pursuit of Law 240/2010
(Gelmini), art. 16 (dedicated to habilitation),
paragraph 3, letter (i). These evaluations to be valid
have to be published integrally and with authors. 20
Over 12628 application here analyzed so far, only
38 times this appeal happened; in both levels of 11A4
it happened 37 times: 18 times for the second level
(5.6% of all applications in its sector) and 19 times
for the first level (17.3% of all applications).
composed of 5 members and to get the
habilitations a candidate must get a positive
evaluation from at least 4 members). Being a
one-step more for recruitment, habilitation
wouldn’t give more efficiency to the system
but could give a contribution for the overall
effectiveness of the recruitment process.
Summing up, the way the habilitation has
worked could be subjected to further analyses
and techniques, included different
approaches21
. These data, however, look to be
quite sound to let say that the indicators of
productivity, especially those measuring
better the quality of research made, let emerge
in quite all the committees here studied, a
reasonable evaluations. Unfortunately very
little can be said about the committees were
the variables here used don’t explain the
outcome of the attainment of habilitation. As a
matter of fact the habilitation tries to
reformulate the human and élite component of
the tribes of scholars (Becher &Trowler,
2001) by allowing in the next steps the local
competitions for permanent positions to have
pre-selected pools of candidates, relying
implicitly to filter out poor candidates,
included those already with permanent
positions (assistant professors and associate
professors) and here define outsiders but
notwithstanding with some years already
committed to research and teaching in tertiary
education.
So for, the change from a collegial into a
more managerial pattern, as seen before in
other contexts (Harle, Muller-Camen, &
Collina, 2004), cannot be considered yet
21
A not suited approach here is the gender one,
which was already addressed for the analysis of
women’s careers in higher education (Bagilhole,
Goode, 2001; Duberleya, Cohen, 2010; van den
Brink, Benschop, Jansen, 2010).
Marini G., Working Paper Cnr-Ceris, N° 03/2014
25
accomplished. If a paradigmatic change will
happen, it would be more probably due to the
combination of the role of habilitation with
other factors such as accountability,
evaluation rationales, and quality assurance.
All of these factors whose effectiveness by
time shall be studied further with more data
and especially with some longitudinal
evidence under investigations.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
26
REFERENCES
Bagilhole B., Goode J. (2001) The
Contradiction of the Myth of Individual
Merit, and the Reality of a Patriarchal
Support System in Academic Careers. A
Feminist Investigation, European Journal of
Women's Studies, 8: 161Becher T., &
Trowler, P. (2001). Academic tribes and
territories. Intellectual enquiry and the
culture of disciplines. Location Needed:
Open University Press.
Bozeman B., Dietz J.S., Gaughan M. (2001)
Scientific and technical human capital: An
alternative model for research evaluation.
International Journal of Technology
Management, 22(7/8), 716-740.
Bourdieu P. (1975) The specificity of the
scientific field and the social conditions of
the progress of reason. Social Science
Information, 14, 19-47.
Cavalli A., Moscati R. (2010) Academic
systems and professional conditions in five
European countries. European Review,
18, 35–53.
Clark B. (1986) The higher education system:
Academic organization in cross-national
perspective. Location Needed: University of
California Press.
Clemente F. (1973) Early career determinants
of research productivity. American Journal
of Sociology, 79 (2), 409-419.
Combes P.P., Linnemer L., Visser M. (2008)
Publish or peer-rich? The role of skills and
networks in hiring economics professors.
Labour Economic, 15, 423–441.
Degli Esposti M., Geraci M. (2010) Thirty
years of higher-education policy in Italy:
Vico’s Ricorsi and beyond? Bulletin of
Italian Politics, 2, 111-122.
Duberleya J., Cohen L. (2010) Gendering
career capital: An investigation of scientific
careers, Journal of Vocational Behavior,
76, 2, 187–197.
Enders J. (2001) A chair system in transition:
Appointments, promotions, and gate-
keeping in German higher education source.
Higher Education, 41, 3-25.
Fassari L. (2009) L’esperienza del Prof. Che
cosa si fa nelle università italiane, Franco
Agenli, Milano.
Ginther D.K., Kahn S. (2004) Women in
economics: Moving up or falling off the
academic career ladder. Journal of
Economic Perspectives, 18(3), 193–214.
Hall B.H., Mairesse J., Turner L. (2007)
Identifying age, cohort, and period effects in
scientific research productivity: Discussion
and illustration using simulated and actual
data on French physicists. Economics of
Innovation and New Technology,
16(2), 159-177.
Harley S., Muller-Camen M., Collina A.
(2004) From academic communities to
managed organisations: The implications for
academic careers in UK and German
universities. Journal of Vocational
Behavior, 64, 329–345.
Kirchmeyer C. (2005) The effects of
mentoring on academic careers over time:
Testing performance and political
perspectives. Human Relations, 58(5),
637–660.
Leahey E., Keith B., Crockett J. (2010)
Specialization and promotion in an
academic discipline. Research in Social
Stratification and Mobility, 28(2), 135-155.
Levin S., Stephan P. (1991) Research
productivity over the life cycle: Evidence
Marini G., Working Paper Cnr-Ceris, N° 03/2014
27
for academic scientists. The American
Economic Review, 81(1), 114-132.
Lissoni F., Mairesse J., Montobbio F., Pezzoni
M. (2011) Scientific productivity and
academic promotion: A study on French and
Italian physicists. Industrial and Corporate
Change, 20(1), 253-294.
Long J.S. (1978) Productivity and academic
position in the scientific career, American
Sociological Review, 43 (6), 889–908.
Long J.S., Fox M.F. (1995) Scientific careers:
Universalism and particularism. Annual
Review of Sociology, 21, 45–71.
Mannella R., Rossi P. (year needed) On the
time dependence of the h-index. Journal of
Infonometrics, 7, 176-182.
McGee R. (1960) The function of institutional
inbreeding. American Journal of Sociology,
65(5), 483-488.
Musselin C. (2004) Towards a European
academic labour market? Some lessons
drawn from empirical studies on academic
mobility. Higher Education, 48(1), 55-78.
Musselin C. (2005) European academic labor
markets in transition. Higher Education,
49,135-154.
Nelken D. (2009) Corruption as governance?
Law, transparency, and appointment
procedures in Italian universities. In F. von
Benda-Beckmann, K. Benda-Beckmann, J.
Eckert (Eds.), Rules of law and laws of
ruling. On the governance of Law (257-
277). Location Needed: Ashgate.
Paris Q. (2005) A mathematical investigation
of “concorsi truccati”. The Italian way of
hiring university professors. Working Paper,
University of California Davis.
Perotti R. (2002) The Italian university
system: Rules and incentives. Rome, Italy:
ISAE.
Pezzoni M., Sterzi V., Lissoni F. (2012).
Career progress in centralized academic
systems: Social capital and institutions in
France and Italy. Research Policy, 41(4),
704-719.
Reale E., Marini G. (2013) La valutazione e i
suoi effetti sull’università: Una
comparazione fra Italia e Francia
(Evaluation and its effects on universities: A
comparison between Italy and France).
Rivista Italiana di Valutazione, Franco
Angeli, 5, 151-170.
Reskin B.F. (1979) Academic sponsorship
and scientists' careers. Sociology of
education, 52, 129-146.
Rossi P. (2012a) Dinamica e prospettive del
reclutamento universitario. Sociologia
Italiana, 0, 159-172.
Rossi P. (2012b) La distribuzione dei
cognomi come strumento per l’analisi
sociale: l’esempio della docenza
universitaria. In A. Addobbati, R. Bizzocch,
& G. Salinero (Eds.), L’Italia dei cognomi
(pp. 203-207). Pisa, Italy: University Press.
Vaira M. (2011) La costruzione della riforma
universitaria e dell'autonomia didattica.
Location Needed: LED Edizioni
Universitarie.
van den Brink M., Benschop Y., Jansen W.,
(2010) Transparency in Academic
Recruitment: A Problematic Tool for
Gender Equality? Organization Studies, 31,
1459-1483.
Marini G., Working Paper Cnr-Ceris, N° 03/2014
28
APPENDIX
Sources. Full detail information can be downloaded as in the first case, or pasted from drop-
down menu in the following two cases.
Matrix [A]: http://cercauniversita.cineca.it/php5/docenti/cerca.php observations are individuals
Matrix [B]: http://abilitazione.miur.it/public/pubblicacandidati.php observations are single
applications (age and gender can be grasped here or from next source)
Vectors of scientific outputs and results of habilitations:
http://abilitazione.miur.it/public/pubblicarisultati.php observations are single applications
List of variables in [F] N=78310:
ID name
SSC (358 variables) dummy (attainment; not attainment)
Who applying insiders; not-applying insiders; applying outsiders
Ruolo positions of the insider in academia
Number of applications made
Scientific area of employee 1-14
Number of articles continuous
Number of citations continuous
H Index continuous
Number of books continuous
Number of chapters and articles continuous
Number of articles in top ranked journals continuous
Pro-veritate judgement dummy
Sex
Age
Marini G., Working Paper Cnr-Ceris, N° 03/2014
29
Figure 1A. Diagram of construction of the database
A
B
C1
[A] ∪ [B]
[B] - [A]
Obs. =
applications
C2
[A] - [B]
Obs. =
applications
Collapsing all
obs. into
individuals
Adding scientific
outputs; sex & age
C3
[A] ∩ [B]
Obs. =
individuals
Adding
attainments of
habilitations
Working Paper Cnr-Ceris
ISSN (print): 1591-0709 ISSN (on line): 2036-8216
Download
www.ceris.cnr.it/index.php?option=com_content&task=section&id=4&Itemid=64
Hard copies are available on request,
please, write to:
Cnr-Ceris
Via Real Collegio, n. 30
10024 Moncalieri (Torino), Italy
Tel. +39 011 6824.911 Fax +39 011 6824.966
segreteria@ceris.cnr.it www.ceris.cnr.it
Copyright © 2014 by Cnr–Ceris
All rights reserved. Parts of this paper may be reproduced with the permission
of the author(s) and quoting the source.